Practice parameter:

Evaluation of the child with

global developmental delay
Report of the Quality Standards Subcommittee of the American
Academy of Neurology and
The Practice Committee of the Child Neurology Society

M. Shevell, MD; S. Ashwal, MD; D. Donley, MD; J. Flint, MD; M.

Gingold, MD; D. Hirtz, MD; A. Majnemer, PhD;
M. Noetzel, MD; and R.D. Sheth, MD.

Published in Neurology 2003; 60:367-380

 The AAN develops these presentation
slides as educational tools for
neurologists and other health care
practitioners. You may download and
retain a single copy for your personal use.
Please contact [email protected] to
learn about options for sharing this
content beyond your personal use.
Objective of the guideline:

To make evidence-based recommendations

concerning the evaluation of the child with a non-
progressive global developmental delay.
Methods of evidence review:
• Literature searches were conducted with the assistance of the
University of Minnesota Biomedical Information Services for
relevant articles published from 1980 to 2000. Databases
searched included MEDLINE, Healthstar, ERIC and CINAHL.
• A bibliography of the 160 articles identified and reviewed for
preparation of this parameter is available at the American
Academy of Neurology Web site (http://www.aan.com/).
• Each article was reviewed, abstracted, and classified by a
committee member. A four-tiered classification scheme for
diagnostic evidence recently approved by the Quality
Standards Subcommittee was utilized as part of this
assessment.
AAN evidence classification
scheme for a diagnostic article
Class I: Evidence provided by a prospective study in a broad
spectrum of persons with the suspected condition, using a “gold
standard” for case definition, where the test is applied in a blinded
evaluation, and enabling the assessment of appropriate tests of
diagnostic accuracy.

Class II: Evidence provided by a prospective study of a narrow

spectrum of persons with the suspected condition, or a well
designed retrospective study of a broad spectrum of persons with
an established condition (by “gold standard”) compared to a broad
spectrum of controls, where test is applied in a blinded evaluation,
and enabling the assessment of appropriated tests of diagnostic
accuracy.
AAN evidence classification
scheme for a diagnostic article
Class III: Evidence provided by a retrospective study where either
persons with the established condition or controls are of a narrow
spectrum, and where test is applied in a blinded evaluation.
Class IV: Any design where test is not applied in blinded evaluation
OR evidence provided by expert opinion alone or in descriptive
case series (without controls).
AAN system for translation of
evidence to recommendations
Translation of evidence to Rating of recommendations
recommendations

Level A rating requires at least A = Established as

one convincing class I study or useful/predictive or not
at least two consistent, useful/predictive for the given
convincing class II studies condition in the specified
population
Level B rating requires at least B = Probably useful/predictive or
one convincing class II study or not useful/predictive for the
overwhelming class III evidence given condition in the specified
population
AAN system for translation of
evidence to recommendations
Translation of evidence to Rating of recommendations
recommendations

Level C rating requires at least C = Possibly useful/predictive or

two convincing class III studies not useful/predictive for the
given condition in the specified
population
U = Data inadequate or
conflicting. Given current
knowledge, test, predictor is
unproven
Introduction
• Developmental disabilities are a group of related chronic
disorders of early onset estimated to affect 5% to 10% of
children.
• Global developmental delay is a subset of developmental
disabilities defined as significant delay in two or more of the
following developmental domains:
– gross/fine motor
– speech/language, cognition
– social/personal
– activities of daily living
• Significant delay is defined as performance two standard
deviations or more below the mean on age-appropriate,
standardized norm-referenced testing.
Introduction
The term global developmental delay is usually reserved for
younger children (i.e., typically less than 5 years of age),
whereas the term mental retardation is usually applied to older
children when IQ testing is more valid and reliable.

Prevalence:
• The precise prevalence of global developmental delay is
unknown.
• Estimates of 1% to 3% of children younger than 5 years of age
are reasonable given the prevalence of mental retardation in the
general population.
• Based on approximately 4 million annual births in the United
States and Canada, between 40,000 to 120,000 children born
each year in these two countries will manifest global
developmental delay.
Clinical Question

What is the diagnostic yield of metabolic and genetic

investigations in children with global developmental
delay?
Analysis of the Evidence
Metabolic testing in children with global
developmental delay
Results
Reference Class N (% patients with abnormal screening)

33 II 151 26% for mild delay; 77% for severe

delay; genetic etiology in 28%, 5%
were metabolic disorders

34 III 1,087 0.6%.; standardized biochemical

screening
Analysis of the Evidence
Metabolic testing in children with global
developmental delay
Results
Reference Class N (% patients with abnormal screening)

35 III 1,568 1.3%; standardized

biochemical screening

7 III 60 63% with all testing but less than 1%

for metabolic testing
36 III 281 5< 5%; nonstandardized evaluation;
etiologic yield of 72% for whole group
Analysis of the Evidence
Metabolic testing in children with global
developmental delay
Results
Reference Class N (% patients with abnormal screening)

28 III 99 <5% even on an indicated basis

37 III 118 13.6% using stepwise rather than

routine screening protocol
Conclusions

Routine screening for inborn errors of metabolism in children

with global developmental delay has a yield of about 1% that
can, in particular situations such as relatively homogeneous
and isolated populations or if there are clinical indicators,
increase up to 5%. When stepwise screening is performed the
yield may increase to about 14%.
Analysis of the Evidence
Cytogenetic studies reporting cytogenetic and
fragile X
Reference Class N Results
% of patients with abnormal results on
cytogenetic studies
41 III 2,757 2.93% (2.61% also had FraX)

42 III 256 3.9%

43 III 274 4.7% (9.1% also had FraX)

44 III 166 5.4%

Analysis of the Evidence
Cytogenetic studies reporting cytogenetic and
fragile X
Reference Class N Results
% of patients with abnormal results on
cytogenetic studies
28 III 99 7.1 %

29 III 120 11.6% (2.3% also had FraX)

7 IV 60 10%

45 IV 170 3.5%
Analysis of the Evidence
Cytogenetic studies reporting on fragile X
prevalence
Reference Class N Results
% of patents with fragile X (FraX)
47 II 1,581 0.7% FraX overall with 1.0% in males,
0.3% in females, and 7.6% in males
with clinically pre-selected criteria
48 II 80 0% females with FraX

49 II 20 0% females with FraX

50 II 278 0.3% females with FraX

Analysis of the Evidence
Cytogenetic studies reporting on fragile X
prevalence
Reference Class N Results
% of patents with fragile X (FraX)
51 II 128 3.9% females with FraX

52 II 50 4.0% females with FraX

53 II 194 4.1% females with FraX

54 II 35 11.4% females with FraX

Analysis of the Evidence
Cytogenetic studies reporting on fragile X
prevalence
Reference Class N Results
% of patents with fragile X (FraX)
55 III 103 3.9% FraX

56 IV 4,940 5.3% FraX

Analysis of the Evidence
Cytogenetic studies testing for Rett syndrome
• Patients with classic Rett syndrome appear to develop normally
until 6 to 18 months of age, then gradually lose speech and
purposeful hand use, and develop abnormal deceleration of
head growth that may lead to microcephaly.
• Seizures, autistic-like behavior, ataxia, intermittent
hyperventilation, and stereotypic hand movements occur in
most patients.
• Rett syndrome is believed to be one of the leading causes of
global developmental delay/mental retardation in females and
is caused by mutations in the X-linked gene encoding methyl-
CpG-binding protein 2 (MECP2). About 80% of patients with
Rett syndrome have MECP2 mutations.
Analysis of the Evidence
Cytogenetic studies testing for Rett syndrome
• The prevalence of Rett syndrome in the general population is
approximately 1 to 3 individuals per 10,000 live births and it
has been estimated that there are approximately 10,000
individuals in the United States with this disorder.
• Currently there are insufficient data to estimate the prevalence
of Rett syndrome variants in milder affected females or in
males.
Analysis of the Evidence
Cytogenetic studies Molecular screening for
subtelomeric chromosomal rearrangements
Reference Class Level of mental N N (% )of patients
retardation with significant
rearrangements

68 I Controls 75 0
Mild 182 1 (0.55)
Moderate/Severe 284 21 (7.39)
69 I Controls 150 0
Unspecified 61 0
Mild 82 0
Moderate/Severe 46 0
Analysis of the Evidence
Cytogenetic studies Molecular screening for
subtelomeric chromosomal rearrangements
Reference Class Level of mental N N (% )of patients
retardation with significant
rearrangements

70 II Unspecified 27 2 (7.4)

71 II Moderate/Severe 29 2† (8.89)
Analysis of the Evidence
Cytogenetic studies Molecular screening for
subtelomeric chromosomal rearrangements
Reference Class Level of mental N N (% )of patients
retardation with significant
rearrangements
72 II IQ< 60 254 13* (5.12)
73 II Unspecified 120 5 (4.17)
Analysis of the Evidence
Cytogenetic studies Molecular screening for
subtelomeric chromosomal rearrangements
Reference Class Level of mental N N (% )of
retardation patients with
significant
rearrangements
74 II Mild 44 0
Moderate/Severe 117 13 (11.11)

75 II Mild 42 0
Moderate/Severe 28 1 (3.57)
Analysis of the Evidence
Cytogenetic studies Molecular screening for
subtelomeric chromosomal rearrangements
Reference Class Level of mental N N (% )of patients
retardation with significant
rearrangements

76 II Unspecified 50 3 (6.0)
77 II Mild 29 3 (10.3)
Moderate/Severe 82 7 (8.5)

78 II Unspecified 250 9 (3.6)

Conclusions
• The accumulated data suggest that cytogenetic studies will be
abnormal in 3.7% of children with global developmental delay,
a yield that is likely to increase in the future as new techniques
are employed.
• In mixed populations (both males and females), a yield of
between 0.3% and 5.3% (average yield of 2.6%) has been
demonstrated for fragile X testing. The higher range of this
yield exists for testing amongst males.
• There is a suggestion that clinical preselection for the fragile X
syndrome amongst males may improve diagnostic testing
beyond routine screening.
Conclusions
• After Down syndrome, Rett syndrome is believed to be the
most common cause of developmental delay in females.
• Although milder variants in females and more severe
phenotypes in males recently have been recognized, estimates
of their prevalence are not currently available.
• Subtelomeric chromosomal rearrangements have been found
in 6.6% (0-11.1%) of patients with idiopathic moderate to
severe developmental delay.
Recommendations
1. Given the low yield of about 1%, routine metabolic screening
for inborn errors of metabolism is not indicated in the initial
evaluation of a child with global developmental delay provided
that universal newborn screening was performed and the
results are available for review. Metabolic testing may be
pursued in the context of historical (parental consanguinity,
family history, developmental regression, episodic
decompensation) or physical examination findings that are
suggestive of a specific etiology (or in the context of relatively
homogeneous population groups) in which the yield
approaches 5% (Level B; class II and III evidence). If newborn
screening was not performed, if it is uncertain whether a patient
had testing, or if the results are unavailable, metabolic
screening should be obtained in a child with global
developmental delay.
Recommendations
2. Routine cytogenetic testing (yield of 3.7%) is indicated in the
evaluation of the child with developmental delay even in the
absence of dysmorphic features or clinical features suggestive of a
specific syndrome (Level B; class II and III evidence).

3. Testing for the fragile X mutation (yield of 2.6%) particularly in the

presence of a family history of developmental delay, may be
considered in the evaluation of the child with global developmental
delay. Clinical preselection may narrow the focus of who should be
tested without sacrificing diagnostic yield. Although screening for
fragile X is more commonly done in males because of the higher
incidence and greater severity, females are frequently affected and
may also be considered for testing. Because siblings of fragile X
patients are at greater risk to be symptomatic or asymptomatic
carriers, they can also be screened (Level B; class II and class III
evidence).
Recommendations
4. The diagnosis of Rett syndrome should be considered in females
with unexplained moderate to severe mental retardation. If
clinically indicated, testing for the MECP2 gene deletion may be
obtained. Insufficient evidence exists to recommend testing of
females with milder clinical phenotypes or males with moderate or
severe developmental delay (Level B; class II and class III
evidence).

5. In children with unexplained moderate or severe developmental

delay, additional testing using newer molecular techniques (e.g.
FISH, microsatellite markers) to assess for subtelomeric
chromosomal rearrangements (6.6%) may be considered (Level
B; class II and class III evidence).
Clinical Question

What is the role of lead and thyroid screening in

children with global developmental delay?
Analysis of the Evidence
Lead Screening

• Lead is the most common environmental neurotoxin. Studies

over several decades have shown a relation between marked
elevations in serum lead levels, clinical symptoms and cognitive
deficits (but not definitively mental retardation).
• Average blood lead levels in the United States have fallen
dramatically from 15µg/dL in the 1970s to 2.7µg/dL in 1991
through 1994.
• It is estimated that there are still about 900,000 children in the
United States between the ages of 1 and 5 years who have blood
lead levels equal to or greater than 10 µg/dL.
Analysis of the Evidence
Lead Screening
• It is unlikely at the present time for a child to have symptomatic
high-level lead exposure that would cause moderate to severe
global developmental delay.
• Low-level lead exposure remains possible, and it has been
estimated that each 10 g/dL increase in blood lead level may
lower a child's IQ by about 1 to 3 points.
• The relation and clinical significance of mildly elevated but
nontoxic levels (i.e., those that do not require medical
intervention) to developmental status remains controversial.
• In a cohort of children (age 12 to 36 months) identified on
routine screening at a urban public hospital, elevated lead
levels (10 to 25 g/dL) resulted in a 6.2- point decline in scores
on the Mental Developmental Index when compared to children
with lead levels below 10 g/dL (class II study).
Analysis of the Evidence
Lead Screening
• In a study of data drawn from the NHANES III, an inverse
relation between blood lead concentration at subtoxic levels
and scores on four measures of cognitive functioning was
demonstrated (class III study).
• Of 72 children referred to a child developmental center with
developmental and/or behavioral problems compared to
controls, a significantly higher distribution of lead
concentrations was demonstrated, with 12% of the sample
possessing a concentration greater than 10 g/dL (class II
study).
• In a study of children drawn from a population at low risk for
lead exposure, 43 children with either developmental delay or
attention deficit hyperactivity disorder did not demonstrate
elevated lead levels compared to controls (class II study).
Analysis of the Evidence
Lead Screening
The recently published guidelines of the American Academy of
Pediatrics, candidates for targeted screening include children 1 to
2 years of age living in housing built before 1950 situated in an
area not designated for universal screening, children of ethnic or
racial minority groups who may be exposed to lead-containing
folk remedies, children who have emigrated (or been adopted)
from countries where lead poisoning is prevalent, children with
iron deficiency, children exposed to contaminated dust or soil,
children with developmental delay whose oral behaviors place
them at significant risk for lead exposure, victims of abuse or
neglect, children whose parents are exposed to lead
(vocationally, avocationally, or during home renovation), and
children of low-income families.
Analysis of the Evidence
Thyroid Screening
• Unrecognized congenital hypothyroidism is a potentially treatable
cause of later developmental delay. Delay in diagnosis and
treatment beyond the newborn period and early infancy has been
clearly linked to later often substantial, neurodevelopmental
sequelae.
• Implementation of newborn screening programs has been
extremely successful in eliminating such sequelae.
• In some countries, where comprehensive newborn screening
programs are not yet in place, congenital hypothyroidism has
been found to be responsible for 17/560 (3.8%) cases of cognitive
delay evaluated in a pediatric neurology clinic (class II Study).
Many of these children also had prominent systemic symptoms.
Conclusions

• Low-level lead poisoning is associated with mild cognitive

impairments but not with global developmental delay.
Approximately 10% of children with developmental delay and
identifiable risk factors for excessive environmental lead
exposure may have an elevated lead level. In the absence of
systematic newborn screening, congenital hypothyroidism may
be responsible for approximately 4% of cases of cognitive
delay.
Recommendations

1. Screening of children with developmental delay for lead toxicity

may be targeted to those with known identifiable risk factors for
excessive environmental lead exposure as per established
current guidelines (Level B; class II evidence).
2. In the setting of existing newborn screening programs for
congenital hypothyroidism, screening of children with
developmental delay with thyroid function studies is not
indicated unless there are systemic features suggestive of
thyroid dysfunction (Level B; class II evidence).
Clinical Question

What is the diagnostic yield of EEG in children with

global developmental delay?
Analysis of the Evidence
EEG
• The vast majority of articles on EEG and global developmental
delay are class IV studies on small cohorts of children with an
already established diagnosis (e.g., sub-acute sclerosing
panencephalitis or progressive myoclonic epilepsy ) that is
often a progressive encephalopathy rather than a static
encephalopathy such as global developmental delay.
• Two class III studies (n=200 children with global developmental
delay) who had EEG have been reported.
• In one study, the EEG did not contribute to determining the
etiology of developmental delay.
• In the second study, 10 of 120 children were found to have
epileptic syndromes. It is likely that all of these children already
had overt seizures and a recognized epilepsy for which an
abnormal EEG result is expected.
Analysis of the Evidence
EEG
• Another class III prospective study of 32 children with
significant developmental dysphasia with or without associated
global developmental delay revealed nonspecific epileptic
abnormalities in 13 of 32 children (40.6%), a finding of unclear
etiologic significance.
• A retrospective class IV study of 60 children with global
developmental delay, 83% of whom had EEG, yielded an
etiologic diagnosis based on the EEG results in 2.0% of the
cohort (specifically one child with ESES–electrographic status
epilepticus during slow wave sleep).
• Although the yield on routine testing is negligible, if there is a
suspected epileptic syndrome that is already apparent from the
history and physical examination (e.g., Lennox- Gastaut
syndrome, myoclonic epilepsy, Rett syndrome), the EEG has
confirmatory value.
Conclusions

• Available data from two class III and one class IV study
determined an epilepsy-related diagnosis in 11 of 250 children
(4.4%). However, the actual yield for a specific etiologic
diagnosis occurred in only 1 patient (0.4%).
Recommendations

1. An EEG can be obtained when a child with global

developmental delay has a history or examination features
suggesting the presence of epilepsy or a specific epileptic
syndrome (Level C; class III and IV evidence).

2. Data are insufficient to permit making a recommendation

regarding the role of EEG in a child with global developmental
delay in whom there is no clinical evidence of epilepsy (Level
U; class III and IV evidence).
Clinical Question

What is the diagnostic yield of neuroimaging in

children with global developmental delay?
Analysis of the Evidence
Neuroimaging
Reference Class N Results
(% patients with abnormal studies)
7 III 60 31.4% (CT, a few had MRI)

28 III 99 27% (CT); 41.2% when done on an

indicated basis vs 13.9% when on a
screening basis
100 III 23 4.3% (CT)

101 III 76 27.6% (CT) with 71.4% of these with

nonspecific atrophy
102 III 37 81% (CT)
Analysis of the Evidence
Neuroimaging
Reference Class N Results
(% patients with abnormal studies)
103 III 79 63% (CT)

45 IV 170 30% (CT); 65.5% (MRI); 19/29

patients who had MRI showed an
abnormality
104 III 224 48.6% ( MRI)

105 III 40 92.5% (MRI)

106 IV 3 100% (MRI) but small number of

patients
Analysis of the Evidence
Neuroimaging
Reference Class N Results
(% patients with abnormal studies)
107 III 13 100% (MRI) but small number of
patients
108 III 21 Amount of abnormal cerebral white
matter on MRI correlated with degree
of mental impairment.
Conclusions

• Available data primarily from class III studies show that CT

contributes to the etiologic diagnosis of global developmental
delay in approximately 30% of children, with the yield
increasing if physical examination findings are present.
• MRI is more sensitive than CT, with abnormalities found in
48.6% to 65.5% of children with global delay with the chance of
detecting an abnormality increasing if physical abnormalities,
particularly cerebral palsy, are present.
Recommendations

1. As the presence of physical findings (e.g., microcephaly, focal

motor findings) increases the yield of making a specific
neuroimaging diagnosis, physicians can more readily consider
obtaining a scan in this population (Level C; class III evidence).

2. If available, MRI should be obtained in preference to CT

scanning when a clinical decision has been made that
neuroimaging is indicated (Level C; class III evidence).
Neuroimaging is recommended as part of the diagnostic
evaluation of the child with global developmental delay (Level
B; class III evidence).
Clinical Question

Are vision and hearing disorders common in children

with global developmental delay?
Analysis of the Evidence
Vision and hearing disorders
• In two class III studies totaling 365 children with global
developmental delay, abnormalities on vision screening were
found in 13% to 25% of children.
• Refractive errors (24%), strabismus (8%), and a number of
organic ocular diseases (8%) were also detected in one of
these reports.
• Supporting these findings is a class IV retrospective review that
estimated the frequency of primary visual sensory impairment
in children with global developmental delay to range between
20% and 50%.
• Appears to be an increased prevalence of additional visual
developmental disability among individuals with syndromes
featuring significant sensory impairment.
Analysis of the Evidence
Vision and hearing disorders
• In one class III study (n=260) with severe global developmental
delay in whom vision and audiologic screening were
performed, 18% of children were found to be deaf.
• Another class III study (n=96) clinically suspected hearing loss
found that 91% had hearing loss as detected by behavioral
audiometry or brainstem auditory evoked response testing.
• Retrospective analysis of legally mandated universal newborn
screening program (53,121 newborns over 4 years)
demonstrated the utility of a two-stage otoacoustic emission
evaluation process in accurately detecting early hearing loss in
a population not amenable to audiometric testing (class II
study).
Conclusions

• Several class III studies have shown that children with global
developmental delay are at risk to have primary sensory
impairments of vision and hearing. Estimates of vision
impairment or other visual disorders range from 13% up to 50%
whereas significant audiologic impairments occur in about 18%
of children based on data in one series of patients.
Recommendations
1. Children with global developmental delay may undergo
appropriate vision and audiometric assessment at the time of
their diagnosis (Level C; class III evidence).

2. Vision assessment can include vision screening and a full

ophthalmologic examination (visual acuity, extra-oculo-
movements, fundoscopic) (Level C; class III evidence).

3. Audiometric assessment can include behavioral audiometry or

brainstem auditory evoked response testing when feasible
(Level C; class III evidence). Early evidence from screening
studies suggest that transient evoked otoacoustic emissions
should offer an alternative when audiometry is not feasible
(Level A; class I & II evidence).
Future Research
Recommendations
1. Further prospective studies on the etiologic yields of various
diagnostic tests need to be undertaken on large numbers of
young children with global developmental delay including control
subjects. These should include newer molecular genetic and
MRI technologies. With this information, prospective testing of
specific evaluation paradigms would be possible.
2. Features (i.e., markers) present on the history and physical
examination at intake need to be identified that will improve
specific evaluation strategies and enhance etiologic yield.
3. The timing of actual testing in children with global developmental
delay needs to be addressed. Specifically, it should be
determined at what age and on what basis one can be certain
that a child has a global developmental delay sufficient to justify
testing as well as at what age the yield from testing will be
optimal.
Future Research
Recommendations
4. Alternative strategies of conducting testing simultaneously or
sequentially need to be critically assessed. This should help
reduce unnecessary testing and provide cost-effective evaluations
and more accurate diagnostic yields.
5. Additional studies are needed to evaluate the role of EEG in a child
with global developmental delay in whom there is no clinical
evidence of epilepsy.
6. Additional studies are needed to better characterize visual and
auditory deficits in children with global developmental delay.
Further investigation of the sensorimotor impairments of children
with global delay are also needed to better determine how early
intervention therapies might improve the overall function of children
who are likely to have multiple needs.
Future Research
Recommendations
7. Issues related to quality of life and social support of families who
have children with developmental delay need further study.
Included in this should be the benefits that medical testing confer
by reducing parental concerns related to determining a specific
etiology and by providing important information regarding
prognosis, genetic counseling, alleviation of parental anxiety, and
planning future educational and treatment needs.
Acknowledgements

The committee thanks the following individuals for their

willingness to review early versions of this manuscript: David
Coulter MD, Child Neurology Society and the American
Association of Mental Retardation; Nancy Dodge MD, American
Academy of Pediatrics, Chair Section on Children with
Disabilities; William Lord Coleman, MD, American Academy of
Pediatrics, Chair Section on Developmental and Behavioral
Pediatrics; Thomas B. Newman, MD, MPH, American Academy
of Pediatrics; Committee on Quality Improvement; Richard
Quint, MD, MPH, Clinical Professor in Pediatrics, UCSF; Carl
Cooley, MD; Constance Sandlin, MD; Clinical Medical Director,
Genzyme Genetics.
To view the entire guideline and additional AAN
guidelines visit:

AAN.com/Guidelines

Published in Neurology 2003; 60:367-380